This research proposal developed from some preliminary observations on a GSH-peroxidase from liver. In contrast to the RBC enzyme and the liver enzyme already reported by others this enzyme is much more stable when purified, does not require activation by preincubation with 25 mM GSH, shows saturation kinetics with GSH (Km equals micron M and does not show saturation with cumene hydroperoxide. We believe that either liver enzyme is somewhat different from the more thoroughly characterized RBC enzyme, that there may be two enzymes in the liver, one cytosolic and one mitochrondrial which have not been sufficiently studied yet to note the differences, or that essentially all RBC enzyme and liver enzyme preparations which have been studied in greater detail have been modified from that native enzyme by organic solvents used in separation of hemoglobin and in purification procedures. Our enzyme is colorless and may be one of a family of selenoenzymes. Since the proposed function of GSH-peroxidase is to protect active membranes from destruction by lipid peroxide formation, a thorough study of both cytosolic and mitochondrial enzymes in many tissues should seem to be in order. GSH-peroxidase is known to be low in some hemolytic conditions of possibly genetic origin, in newborn, and in selenium deficiency. The full pathological implications in all tissues are not yet known, as vitamin E can in part prevent expression of GSH-peroxidase deficiency.